publication

MR-only brain radiotherapy: Dosimetric evaluation of synthetic CTs generated by a dilated convolutional neural network

Dinkla, AM, Wolterink, JM, Maspero, M, Savenije, MHF, Verhoeff, JJC, Seravalli, E, Isgum, I, Seevinck, PR, van den Berg, CAT

DOI: http://dx.doi.org/https://doi.org/10.1016/j.ijrobp.2018.05.058

International Journal of Radiation Oncology Biology Physics

Abstract

Purpose
This work aims to facilitate a fast MR-only workflow for radiotherapy of intracranial tumors. Here, we evaluate whether synthetic computed tomography (sCT) images generated with a dilated convolutional neural network (CNN) enable accurate magnetic resonance (MR)-based dose calculations in the brain.

Methods
We conducted a retrospective study on 52 patients with brain tumors who underwent both CT and MR imaging for radiotherapy treatment planning. To generate the sCTs, a T1-weighted gradient echo MR sequence was selected from the clinical protocol for multiple types of brain tumors. sCTs were created for all 52 patients with a dilated CNN using two-fold cross validation, in each fold 26 patients were used for training and the remaining 26 for evaluation. For each patient, the clinical CT-based treatment plan was recalculated on sCT. We calculated dose differences and gamma pass rates between CT- and sCT-based plans inside body and planning target volume (PTV). Geometric fidelity of the sCT and differences in beam depth and equivalent path length were assessed between both treatment plans.

Results
sCT generation took 1 minute per patient. The mean absolute error (MAE) over the patient population of the sCT within the intersection of body contours was 67±11 Hounsfield Units (HU) (±1SD, range: 51-117 HU), and the mean error (ME) was 13±9 HU (±1SD, range:-2 to 38 HU). Dosimetric analysis showed mean deviations of 0.00±0.02% (±1SD, range:-0.05-0.03) for dose within the body contours and -0.13±0.39% (±1SD, range:-1.43-0.80) inside the PTV. Mean γ1mm/1% was 98.8±2.2% for doses larger than 50% of the prescribed dose.

Conclusions
The presented dilated CNN generated sCTs from conventional MR images without adding scan time to the acquisition. Dosimetric evaluation suggests that dose calculations performed on the sCTs are accurate, and can therefore be used for MR-only intracranial radiotherapy treatment planning.